Inactivation of the gene encoding neurofibromin results in specific and predictable congenital heart malformations in the mouse embryo. These defects involve the endocardial cushions and the myocardium and lead to mid-gestation cardiac failure and fetal demise. In addition, abnormalities of neural crest derivatives, including dorsal root and sympathetic ganglia are present. Neural crest cells also contribute to the heart and are required for proper development of the outflow tract, a region affected in Nf1 mutant embryos. It is unknown whether Nf1 deficiency in a single population of cells results in the complex phenotype through non-cell autonomous effects, or whether Nf1 exerts critical functions in several tissues during development. For example, the myocardial defect of Nf1 mutant embryos may be secondary to altered hemodynamics and outflow obstruction due to enlarged endocardial cushions, a condition that may be related to neural crest abnormalities. If Nf1 serves critical functions in several distinct tissues, does it do so by regulating similar intracellular signaling pathways? We have shown that the endocardial cushion defect of Nf1 mutant embryos is due to up- regulation of Ras activity, and I predict that blocking Ras activation will improve or correct this aspect of the congenital heart disorder. Here, we will further define the Ras pathways regulated by neurofibromin during cardiac development. There is evidence that N-ras is required for expression of the learning disorder manifest by heterozygous Nf1 deficient mice. We will test whether N-ras is required for expression of the homozygous cardiac defects. Finally, we will examine the effects of specific Nf1 ablation in neural crest or myocardium to address what aspects of neurofibromin deficiency are due to cell autonomous defects. Specific aims are: 1. To elucidate the intracellular Ras pathway affected by neurofibromin deficiency during endocardial cusion formation using an explant culture system, and to test whether Ras activation can replace the requirement for TGFbeta during this process. 2. To assess the contribution of n-ras to the Nf1 mutant cardiac phenotype by breeding Nf1 mutant mice that also lack N-ras. 3. To inactivate the Nf1 gene selectively in neural crest cells and separately in myocardial cells using tissue specific promoters and a Cre/lox strategy.